I. Introduction

In the realm of macroeconomic analysis, the intricate interplay between inflation and economic growth remains a subject of intense scrutiny and debate, especially in the context of emerging market economies. These economies, characterized by rapid development and dynamic economic landscapes, present unique challenges and opportunities for macroeconomic stabilization policies. These policies aim to foster sustainable growth while managing inflation. Among these economies, China stands out as a prominent example, having undergone significant economic transformations accompanied by fluctuating inflation levels. This study delves into the complex relationship between inflation and economic growth in China, an economy that not only plays a pivotal role in the global economic arena but also exhibits high volatility in its growth and inflation rates. The necessity of this investigation is underscored by the cyclical nature of inflation within the business cycle, typically marked by phases of economic prosperity followed by inflationary downturns, recessions, and disinflationary expansions. This pattern raises pivotal questions about the transmission channels through which inflation impacts economic growth and the variations that exist across different countries or regions. The recent COVID-19 pandemic has added layers of complexity to this dynamic, introducing unprecedented levels of uncertainty and instability in both inflation and economic growth. In countries like China, which are already characterized by high volatility in growth and inflation rates, the pandemic has exacerbated these challenges, making the need for a nuanced understanding of the inflation-growth relationship even more critical. This paper uses the latest data and combines both threshold and Vector Autoregression (VAR) models to conduct a comprehensive analysis of the relationship between inflation and economic growth. Set against the backdrop of significant fluctuations in China’s inflation and economy due to the COVID-19 pandemic, this study holds substantial contemporary relevance.

Our study utilizes annual data from China, covering the period from 1987 to 2021, to explore the complexities of the relationship between inflation and economic growth. We employ threshold models and Vector Autoregression (VAR) models for this analysis. The threshold model, known for its effectiveness in analyzing nonlinear relationships between variables, helps us investigate whether there is a specific inflation threshold in China. Beyond this threshold, we hypothesize that the impact of inflation on economic growth may change significantly. Conversely, the VAR model enables us to examine the dynamic interactions and mutual influences between inflation and economic growth. By using these methods, our study aims to provide a more comprehensive understanding of the relationship between these two critical economic factors, offering deeper insights into their interconnectedness.

The empirical findings of our study indicate a significant threshold effect in the relationship between inflation and economic growth in China. Specifically, when inflation exceeds 2%, it adversely affects economic growth. In contrast, inflation below this threshold appears to positively influence growth. At low inflation rates, consumer and business confidence is bolstered, leading to increased consumption and investment, which in turn stimulates economic demand. Low inflation enhances the propensity to purchase, particularly in long-term consumer goods like durable items and real estate. It also encourages businesses to expand production and investment, creating job opportunities. Furthermore, low-interest rate policies, often associated with low inflation, reduce borrowing costs, promoting further investment and consumption. On the other hand, high inflation rates devalue currency, increase economic uncertainty and risk, impact long-term planning and investment decisions, and lead to distorted resource allocation, all of which constrain economic growth. Additionally, our analysis using the Vector Autoregression (VAR) model reveals a unidirectional causal relationship between inflation and economic growth. It also shows the diminishing contribution of inflation to economic growth over time, a key insight from our quantitative analysis. These findings have undergone rigorous robustness checks, enhancing the reliability of our research. However, the methodological approach of this study has certain limitations from an economic perspective. The annual time series data used spans from 1987 to 2021. While this is adequate for analyzing long-term economic trends, the granularity of annual data may not capture the nuanced effects of economic fluctuations and policy shifts within shorter periods. For a comprehensive understanding of economic cycles, future research should consider incorporating quarterly or monthly data. This would provide a higher temporal resolution, allowing for a more sensitive and accurate analysis of short-term fluctuations and their impact on long-term trends. Such an approach would offer a more robust foundation for economic policy formulation.

This research confirms the existence of an inflation threshold level in emerging market economies, significantly enriching the existing academic literature. This concept, which has evolved to reflect a deeper understanding of the inflation-economic growth relationship, marks a departure from early economic research that primarily relied on a linear assumption. In this traditional view, inflation was seen as having a consistent, unidirectional effect on economic growth, regardless of its level. Early economic theories varied in their perspectives on inflation's impact. Tobin (1965) highlighted the positive role of monetary supply in fostering economic growth, while Sidrauski (1967) and Mundell (1965) pointed out the adverse effects of inflation, such as inefficient resource allocation and economic instability. Recent studies, however, suggest a nonlinear relationship. Bruno and Easterly (1998) argued that high inflation could lead to economic instability and slow growth, potentially triggering self-fulfilling inflationary cycles or reducing investment and consumption due to increased production costs and decreased purchasing power. Wu and Zhang (2012) used a dynamic panel smooth transition regression (PSTR) model to study China’s economy, finding that low inflation rates positively affect economic growth, but as inflation increases, its impact becomes mixed. Further supporting this nonlinear perspective, Phiri (2018) identified a specific inflation threshold in South Africa, with inflation below 5.4% positively affecting economic growth and higher levels having a negative impact. Ngoc (2020) also observed asymmetric influences of inflation on Vietnam’s economic growth, indicating that inflation negatively and unevenly affects long-term growth. These findings collectively challenge the traditional linear assumption, suggesting that the relationship between inflation and economic growth is more complex and varies across different economic contexts.

This study confirms the existence of a threshold level of inflation in an emerging market economy, aligning with previous research such as Khan and Senhadji (2001), Eggoh and Khan (2014), Mubarik and Riazuddin (2005), Fakhri (2011), and Vinayagathasan (2013). Unlike traditional linear models, our threshold model captures nonlinear effects, indicating that the impact of inflation on economic growth may significantly change when it reaches a certain level. This approach provides a comprehensive understanding of the inflation-economic growth relationship, revealing potential threshold effects. However, the 2% inflation threshold identified for the Chinese economy in this study is notably lower than those found in other studies. For instance, Khan and Senhadji (2001) suggested a threshold of 11% for developing countries, Mubarik and Riazuddin (2005) found 9% for Pakistan, and Fakhri (2011) estimated 13% for Azerbaijan. Vinayagathasan (2013) identified a 5.43% threshold across 32 Asian countries, noting that inflation rates above this level adversely affect economic growth. The relatively low inflation threshold in China can be attributed to its unique economic structure and development strategy, including a deep dependence on manufacturing and exports, rapid industrialization and urbanization processes, and a high emphasis on price stability. These factors collectively shape the sensitivity of the Chinese economy to inflation, leading to the need for a delicate balance between maintaining economic growth and controlling inflation. Firstly, as the “world’s factory,” China’s reliance on manufacturing and exports compels policymakers to maintain a low inflation rate to keep its goods competitively priced on the global market (Lardy, 2002). This dependence means that fluctuations in external market demand can significantly impact the Chinese economy, while maintaining price stability is key to attracting foreign direct investment. Secondly, China’s rapid industrialization and urbanization have driven substantial demand for infrastructure and housing, necessitating governmental measures to keep inflation low to reduce financing costs and support continued economic expansion (Yao and Zhang, 2001). Lastly, the Chinese government places a high priority on price stability, viewing it as crucial for maintaining consumer and business confidence and promoting healthy economic growth. Through effective macroeconomic policies, the government strives to control inflation to avoid its potential negative impacts on economic growth and social stability (Hu and Reardon, 2014). Additionally, the proactive inflation control policies of the People’s Bank of China and the country’s sensitivity to international capital flows, export competitiveness, and foreign investment due to its open economy status may contribute to this lower threshold. Our study also employs the VAR model to examine the dynamic interplay between inflation and economic growth in China. Understanding the formation of inflation expectations and underlying drivers is crucial. In this context, the study by Arce­Alfaro and Blagov (2023) is pertinent. Using U.S. data and a stochastic volatility-in-mean structural VAR model, they found that monetary policy uncertainty reduces both inflation expectations and actual inflation, a link that has persisted since the global financial crisis. Furthermore, Finck and Tillmann (2022) explored the effects of global and domestic shocks on inflation dynamics in Asia, highlighting the significant role of global shocks in influencing the inflation and output dynamics of emerging Asian economies. Other notable studies employing VAR models include Wang (2009), who found a bi-directional Granger causality between inflation and economic growth in China, and Kim and Lee (2014), who revealed a bi-directional long- and short-term causal relationship between these variables in South Korea. Our study’s identification of a specific inflation threshold in China contributes valuable insights and policy implications for monetary policymakers, particularly in the context of emerging market economies.

The structure of this paper is organized as follows: Section 2 examines the relationship between inflation and economic growth in China. Section 3 introduces the data and variables utilized in this study, and details the model and theoretical framework employed. Section 4 presents the results of the empirical analysis. Section 5 concludes the paper, summarizing key findings and implications.

II. China’s Inflation and Economic Growth

From 1987 to 2021, China’s economy experienced rapid expansion, a trend that is demonstrated in Figure 1 through the data on the real GDP growth rate (denoted as “GDP”). During the same period, inflation, represented by the Consumer Price Index (CPI) growth rate (indicated by “CPI”), exhibited significant volatility. From Figure 1, we can observe Figure 1 does not offer a “clear” depiction of the relationship between economic growth and inflation rate, making the clarification of the specific relationship between the economic growth rate and inflation rate the primary objective of our study. Notably, in the early 1990s, China experienced exceptionally high inflation rates, peaking at around 25% in 1994. This period presents diverse perspectives on the impact of inflation on economic growth. Some researchers have suggested that elevated inflation during the early 1990s may have contributed to economic instability in China, subsequently slowing economic growth. Lardy (1998) provided a comprehensive analysis of the challenges and processes involved in China’s economic reforms. He emphasized that pronounced inflation during the early 1990s emerged from a combination of lenient monetary policies and excessive administrative interventions. This inflation spike created business uncertainty and misdirected resources, negatively impacting economic growth. Similarly, Brandt and Zhu (2000) examined the relationship between growth and inflation in China during its economic reform period. They argued that fiscal decentralization in local governments, coupled with inconsistent monetary policies, were pivotal in causing an inflation surge in the early 1990s. This inflationary environment hindered economic growth by disrupting pricing mechanisms, leading to inefficient resource distribution.

In contrast, some studies have contended that the relationship between inflation and economic growth in China during this era was more nuanced, implying that other variables, like increased investment and exports, may have significantly influenced economic performance. Christoffersen and Doyle (1998) examined China’s trajectory from elevated inflation to stable growth. They underscored that despite the adverse effects of inflation, China sustained strong growth in the early 1990s, largely owing to a boost in exports and foreign investments. Kraay (2000) examined the surge in household savings in China and its economic implications. He argued that even during high inflation, increased household savings injected critical capital for investments, thereby underpinning economic growth. Additionally, Rawski (1994) probed China’s industrial reforms and their impact on the economy. He highlighted that despite inflation challenges, the nation continued to experience considerable growth, owing to industrial reforms and improvements in production efficiency. Nevertheless, further research and in-depth analyses are essential to better comprehend the relationship between inflation and economic growth, thereby facilitating the formulation of appropriate economic policies.

The analysis of Figure 2 highlights the Chinese government’s response to the economic shock caused by the COVID-19 pandemic. Proactive monetary policies were implemented, notably increasing the money supply to stimulate the economy. In 2020, the annual growth rate of the M2 money supply surged to 11%, a move aimed at alleviating financial pressures on businesses and consumers by enhancing liquidity. By 2021, as the economy began to recover, the growth rate of the money supply moderated to around 8%. In terms of price levels, the initial stages of the pandemic witnessed a rise in prices due to supply-demand imbalances, with the Consumer Price Index (CPI) climbing to 5.4% at the end of 2019 and the beginning of 2020. However, as the pandemic prolonged and economic activities slowed, the CPI experienced a significant drop towards the end of 2020, nearing 0. This decline likely reflected a sharp decrease in demand and market price adjustments. In 2021, with the economy’s gradual recovery and market stabilization, the inflation rate modestly rebounded to about 2%. Figure 3 shows the trajectory of the Purchasing Managers’ Index (PMI) in China, a key indicator of economic activity. The PMI experienced a sharp decline in 2020 due to the pandemic. However, with the global pandemic situation improving and the resumption of economic activities in other countries, China’s exports increased significantly in 2021, leading to a substantial recovery in the PMI. As a major global manufacturing hub, China benefited from the rebound in global demand, particularly in sectors like electronics, medical supplies, and home goods. The non-manufacturing PMI, in contrast, displayed greater volatility, reflecting the diverse and complex challenges faced by the service industry. The insights from Figures 2 and 3 reveal a close relationship between inflation rates and economic growth. The significant downturn in economic activities at the onset of 2020 was mirrored by a decrease in the inflation rate, while the economic recovery in 2021 was paralleled by a moderate rise in inflation. However, despite the economic recovery signs in 2021, the inflation rate remained relatively low, possibly due to the gradual restoration of supply chains and ongoing market adjustments.

Ⅲ. Data and Methodologies

1. Variables and Data

The main goal of this study is to examine the relationship between inflation and GDP growth rates in China. To do so, we used annual time series data spanning 1987 to 2021 in China, sourced exclusively from the World Bank and Federal Reserve Economic Data (FRED). The main economic variables examined in this study include China's Real GDP expressed in 2017 U.S. dollars (notated as “GDP”) and the annual inflation rate measured by the Consumer Price Index (notated as “CPI”). We also included important macroeconomic variables that could potentially affect the growth rate. Furthermore, based on the research by Rey (2015) and Miranda-Agrippino and Rey (2020), which indicates that GDP growth and inflation in open economies are significantly influenced by global factors, we have included variables representing global influences in our model. Consequently, the control variables incorporated into our model consist of Foreign Direct Investment to China (denoted as “FDI”), the Chicago Board Options Exchange Volatility Index: VIX (denoted as “VIX”), the real interest rates in the United States (denoted as “RIR”), and China’s real general government final consumption expenditure expressed in domestic currency (denoted as “GEXP”).

2. Quantitative Regression Model: Threshold Approach

The threshold regression model proposed by Khan and Senhadji (2001) is employed to analyze the threshold level of inflation. It estimates the nonlinear impact of inflation on economic growth in China, both below and above the identified threshold. We initiated our analysis by conducting unit root tests on the observed variables. The results indicated that the data are non-stationary series; however, after taking the first difference of the variables, they were found to be stationary. Additionally, the Johansen cointegration test revealed two cointegration relationships among the variables. This implies that although short-term deviations may exist among these variables, they eventually converge to their long-term equilibrium relationships.

Next, the basic regression model is represented by the following equation:

In Equation (1), t represents the time point (year) at which the data is collected. Based on the above basic model, we build a threshold model, following Khan and Senhadji (2001), with reference to the model specifications of Mubarik and Riazuddin (2005) and Fakhri (2011). The inflation threshold effect on economic growth is represented by the following equation.

where vector Z_t includes three control variables from equation (1)—FDI_t, VIX_t, RIR_t and GEXP_t and D_t is a dummy variable which is defined by the following equation:

The parameter k represents the threshold inflation level. As indicated by Mubarik and Riazuddin (2005) and Frimpong and Oteng-Abayie (2010), the threshold value k signifies a turning point; when inflation reaches or surpasses k, its influence on economic growth experiences a noteworthy shift. Within this model, α₁ and α₂ are regression coefficients, detailing the impacts of low and high inflation on economic growth, respectively. If the inflation rate stays below k, then α₁ is the dominant influencing factor. However, once inflation exceeds k, the cumulative effects of α₁ and α₂ (α₁ + α₂) dictate inflation’s effect on growth. To determine the optimal threshold k, researchers conduct regression analyses across various k values, aiming to identify the value k that optimizes the R-squared value (or to minimize the Residual Sum of Squares). This paper references the studies of Mubarik and Riazuddin (2005) and Fakhri (2011), adopting the criteria of maximizing R2 and minimizing the residual sum of squares (RSS) for selecting the optimal threshold k. Additionally, based on the research by Hansen (2000), the optimal threshold is determined through minimizing the sequence of residual sum of squares during threshold estimation. Hansen discussed the application of Least Squares (LS) estimation in threshold regression models and detailed how to use LS estimation to find the threshold that minimizes RSS. This process involves trying different thresholds on a given dataset and selecting the one that minimizes the total residual sum of squares of the model as the optimal threshold. The optimal threshold determined by minimizing the residual sum of squares has the characteristic of significantly impacting economic growth at that threshold level. This means that the effect of inflation on economic growth differs below this threshold compared to when it exceeds this threshold. Further, Khan and Senhadji (2001) provided an exhaustive breakdown of the estimation approach and computational methodologies underlying this model. Their study is essential for research in this area.

3. Granger Causality Test and Variance Decomposition Analysis

(1) VAR model

The decision to prioritize VAR-based tests over basic ordinary least squares (OLS) analysis stems from the VAR model’s ability in identifying complex relationships that may elude traditional methodologies. The VAR-based tests provide a nuanced understanding of how variables interact over time, including their potential feedback loops, laying the groundwork for subsequent analyses. Following this detailed examination, we performed a core OLS analysis. This phase facilitates a thorough assessment of the linear relationships among the variables, thereby deepening our understanding of the link between inflation and economic growth.

In this study, we conducted the Granger causality test and variance decomposition analysis using the VAR model, along with stationary tests, before applying the regression to the threshold model (2). This approach was adopted primarily because these methods supplement our main model in various ways. The VAR model is widely used to examine the relationships among macroeconomic variables, such as GDP, inflation rate, and interest rate. It is constructed based on the statistical properties of data. In this system, each endogenous variable is considered in terms of the lag values of all other endogenous variables, thereby building the model. This approach extends the univariate autoregressive model into a “vector” autoregressive model, encompassing multiple time series variables.1 First, the VAR model captures lagged effects between variables, a crucial aspect when analyzing the impact of inflation rate changes on economic growth; these macroeconomic variables often exhibit certain lagged responses. Second, the VAR model addresses autocorrelation error terms by managing the potential issues of serial correlation. However, based on the study by Lütkepohl (2005), the VAR model necessitates a sufficient sample size to ensure the accuracy of its estimates, a criterion our dataset unfortunately did not satisfy. Therefore, despite the VAR model’s significant advantages in multivariate time series analysis, such as its capacity to capture the dynamic interrelations among variables, its limitations are particularly pronounced with an insufficient sample size. These limitations include instability in the estimation results and potential biases. Given these constraints associated with the VAR model, we have pivoted to the simpler threshold regression model in our subsequent analysis. The threshold regression model is capable of more effectively capturing the data’s nonlinear characteristics and dynamic shifts with a smaller sample size. Furthermore, the threshold model allows us to divide the sample into different regions or phases based on predefined values of variables, thereby enabling a more precise analysis of the relationships among variables across various economic conditions or policy settings.

(2) Granger causality test

The Granger causality test enables the identification of the causal direction between the main variables, i.e., whether one variable can explain the changes in the others, thereby revealing the causal relationships among them. Through variance decomposition analysis, the explanatory power of each variable on the overall volatility can be quantified, revealing the contribution of each variable to the system’s volatility. This helps us understand the system’s structure and the relative importance of the variables.

Specifically, the Granger causality test was employed to examine the causal relationship between economic growth and inflation. The null hypothesis of the test is that there is no Granger causality between the two variables, indicating that one variable’s past values do not provide information about the future values of the other. The alternative hypothesis suggests Granger causality, implying that one variable’s past values can be used to predict the future values of the other. When conducting the Granger causality test, we first established a VAR model to determine the optimal lag order. We then estimated two separate autoregressive models for each variable: one included its own lagged values, while the other included the lagged values of both variables. Subsequently, we employed the F-test to compare the predictive ability of the two models. If the F-test indicates that the joint model with both variables provides better predictions than the individual models, we can reject the null hypothesis and conclude that Granger causality exists between the variables. Conversely, if the F-test suggests that the joint model does not offer better predictions than the individual models, we fail to reject the null hypothesis and conclude the absence of Granger causality between the variables.

(3) Variance decomposition

Variance decomposition partitions a variable’s total variance into components attributable to various sources or factors. It is commonly used in time series analysis and other statistical modeling to determine the relative importance of different variables or factors in explaining the behavior of a target variable. In a time series analysis, variance decomposition is often employed to decompose the target variable’s variance into portions attributable to its own lagged values and those of other related variables. This helps identify the relative importance of these different sources of variation in the behavior of the target variable. Depending on the context and type of model used, different methods are employed to conduct variance decomposition. One commonly used approach involves calculating the contributions of different factors to the target variable’s total variance using the variance-covariance matrix of model errors. This method was employed in this study as well.

1)See Luo et al. (2012) for more details.

IV. Estimation Results

1. Stationary Test Results

We initially conducted unit root tests for all variables in regression models (1) and (2). In their original form, the time series of all variables exhibited non-stationarity, indicating the presence of unit roots. However, considering their original differences, we successfully transformed these variables into stable time series, eliminating trends and seasonal fluctuations present in the original series. The Johansen cointegration test was employed, based on the VAR model, to test for cointegration among the time series data. In the Johansen test, the calculated cointegration statistic is compared with the corresponding critical values. If the cointegration statistic exceeds the critical value, we can reject the null hypothesis of no cointegration and conclude that a cointegration relationship exists. According to the results in Table 1, the cointegration statistics reject the null hypothesis at the 10% significance level when considering four or fewer longrun cointegration relationships. However, when considering five long-run cointegration relationships, the obtained cointegration statistic fails to reject the null hypothesis. Therefore, we conclude that four long-run cointegration relationships exist among the variables. This implies that the variables share a common long-term trend over time. While short-term deviations may occur, they eventually revert to a long-term equilibrium state.

2. Granger Causality Test Results

The results of the Granger causality tests based on the VAR model are presented in Table 2, indicating the presence of Granger causality between the inflation rate and economic growth. Specifically, the results show that the inflation rate has a causal impact on economic growth, implying that changes in the inflation rate affect economic growth. However, economic growth does not have a causal relationship with the inflation rate, implying that changes in economic growth do not directly lead to changes in the inflation rate. Additionally, we found that FDI has a causal impact on economic growth, implying that changes in FDI influence economic growth. However, economic growth also does not directly affect FDI. The causal relationships between U.S. interest rates, VIX and Chinese government spending with economic growth yielded similar results.

3. Variance Decomposition Analysis Results

The results of the variance decomposition analysis, presented in Table 3, indicate a trend where the contribution of inflation to China's economic growth initially increases and then decreases. This finding reveals a complex and dynamic relationship between the inflation rate and economic growth. In the initial stages of rising inflation, its positive impact on economic growth is primarily manifested through stimulating consumption and investment to increase economic activity (Tobin, 1965), as well as encouraging economic entities to increase spending by reducing the real burden of debt (Fisher effect). However, as the inflation rate continues to rise, its negative impacts begin to emerge, including decreased consumer and investor confidence due to increased economic uncertainty (Bruno and Easterly, 1998), and reduced efficiency in resource allocation caused by price instability (Fischer, 1993). Moreover, the changing relationship between inflation and economic growth may relate to inflation reaching a threshold beyond which the economy can sustain growth (Khan and Senhadji, 2001). When the inflation rate exceeds this threshold, the stimulating effects on economic growth are offset by negative effects such as increased uncertainty and rising costs. These findings underscore the importance of considering the multifaceted impact of inflation on economic growth when formulating macroeconomic policies and the necessity of implementing moderate inflation control measures to promote economic stability and growth. The observed trend of FDI’s contribution to economic growth, which first decreases then increases, can be attributed to adjustments in China’s economic structure and industrial upgrading. The initial dependence on foreign capital diminishes as policies are optimized and markets further open, enhancing the quality and efficacy of FDI, thereby making a greater contribution to economic growth (Zhang and Daly, 2011). Concurrently, the increasing contribution of the VIX index to China’s economic growth over time may reflect market participants' growing concerns about future uncertainties. While such uncertainties may suppress economic activities in the short term, in the long run, they encourage both enterprises and the government to strengthen risk management and adjust economic structures, potentially indirectly promoting the stability of economic growth (Baker et al., 2016). Additionally, the gradual decline in the contribution of U.S. real interest rates to China’s economic growth might reflect the diminishing direct impact of U.S. monetary policy as China’s economy matures and the influence of international financial markets changes (Rey, 2015). The decreasing contribution of Chinese government spending to economic growth could indicate a decline in the efficiency of government expenditures, especially in the context of diminishing marginal effects of investments in infrastructure and public services, necessitating a focus on quality and efficiency in government spending as the economy shifts from rapid to high-quality development (Barro, 1990).

4. Basic Model Regression Results

The results of the least squares regression based on model equation (1) are presented in Table 4. The results show that the inflation rate negatively impacts economic growth, indicating a negative correlation between increasing inflation rates and decreasing economic growth. This negative effect persists even after controlling for other variables. One possible explanation is the cost-push effect. As the inflation rate increases, certain costs, such as production, labor, and raw material, may increase. This can affect firm profitability and investment decisions, thereby dampening economic growth.2 Another possible explanation is the expectation effect. An increase in the inflation rate may trigger an increase in expected inflation. Consequently, businesses and individuals may react by accelerating consumption, advancing investments, and purchasing tangible assets. This behavior may provide a short-term boost to economic growth. However, the premature release and offsetting of consumption and investment may constrain long-term economic growth, resulting in a negative impact of inflation rate on economic growth.3 In addition, both higher CPI inflation and higher VIX (global stock market volatility) have a negative impact on the growth rate. Generally, external foreign shocks are transmitted to the domestic region with a degree of price uncertainty, such as exchange rate pass-through. If local (domestic) inflation is anchored at a lower level, as we found, these external impacts will not be amplified, exacerbating domestic economic activities.

5. Threshold Model Regression Results

In our study, we employed the threshold model outlined in equation (2) to investigate the relationship between economic growth and inflation, this exploration conducted across a range of potential inflation thresholds, from k=1(%) to k=16(%). According to Mubarik and Riazuddin (2005), Frimpong and Oteng-Abayie (2010), and Fakhri (2011), the parameter k represents the threshold level of inflation, possessing the following characteristics: When the relationship between economic growth and inflation is: (i) low inflation: α1; (ii) high inflation: α1 + α2. High inflation implies that when inflation estimates are significant, (α1 + α2) are added simultaneously to observe their impact on growth, which would be the threshold level of inflation. The optimal value of k is obtained by estimating regressions for different k values, which are arbitrarily chosen in ascending order (i.e., 1, 2, 3, 4, 5, etc.), by finding the value that maximizes the R-squared (R²) or minimizes the sum of squared residuals (RSS) of the corresponding regression. The lack of knowledge about the optimal number of threshold points and their values complicates estimation and inference. Although this procedure is widely accepted in empirical literature, it is tedious because multiple regressions must be estimated. Khan and Senhadji (2001) discuss the details of the estimation procedure and computational methods. The findings, as presented in Table 5, reveal that the most satisfactory threshold level of inflation, k, is achieved at 2%. Therefore, when the inflation rate is below 2%, the estimated effect of the inflation rate on economic growth is indicated by the estimated coefficient in Equation (2). However, when the inflation rate exceeds 2%, the estimated effect of the inflation rate on economic growth is the sum of and the interaction term between the dummy variable and the inflation rate In the analysis of the impact of inflation on economic growth in China, this study identifies a critical inflation threshold rate of 2%, a finding that holds significant implications for understanding and formulating macroeconomic policies. From a long-term perspective, when the inflation rate exceeds this threshold, it negatively impacts the Chinese economy, manifested in increased capital costs, reduced purchasing power for consumers, and heightened economic uncertainty, all of which suppress economic growth. Conversely, an inflation rate below 2% indicates a relatively stable price level, conducive to healthy economic development (Svensson, 1997). The inflation threshold identified in this study closely relates to the concept of Non-Accelerating Inflation Rate of Unemployment (NAIRU) within New Keynesian economics. NAIRU represents the lowest unemployment rate an economy can sustain without triggering an increase in the inflation rate (Gordon, 1997). Viewing the 2% inflation rate as a target reflects an attempt to balance economic growth and reduce unemployment without promoting inflation. Many central banks worldwide, including the European Central Bank and the Federal Reserve System of the United States, have set their inflation target at 2%. This target level is deemed to encourage investment and consumption while maintaining the stability of currency value, thereby averting the risks of economic deflation (Bernanke et al., 1997). Thus, adopting a 2% inflation rate as a target is not only theoretically sound but has also been widely accepted in practice. The findings of this paper emphasize the need to consider the relationship between the inflation rate and economic growth in the formulation and adjustment of macroeconomic policies, as well as the importance of stabilizing the price level. Moreover, the 2% inflation threshold provides policymakers with a clear target, guiding the formulation of economic policies towards achieving long-term economic stability and growth.

Specifically, when the inflation rate is below 2%, it has a positive effect of 0.035% on economic growth. When the inflation rate is low, it encourages consumers and businesses to anticipate future price increases, leading to more immediate consumption and investment, thereby stimulating economic demand. Firstly, from the consumer perspective, a stable inflation rate boosts consumer confidence, prompting them to make long-term consumption plans, such as purchasing durable goods and real estate, which directly stimulates economic activity. Then, from the business perspective, a low inflation environment helps companies maintain or increase their profit margins, spurring them to expand production and investment, create jobs, and drive economic growth. Additionally, low-interest rate policies under low inflation reduce borrowing costs, making it easier for individuals and businesses to obtain loans, stimulating investment and consumption, and thereby promoting economic growth. Finally, a low inflation rate typically signifies price stability, high employment rates, and stable currency value, creating a favorable environment for long-term economic planning and sustained growth. Conversely, when the inflation rate exceeds 2%, it has a statistically significant negative impact on economic growth with an estimated volume of -0.002%, indicating a clear nonlinear relationship between the inflation rate and economic growth. A rapid increase in the inflation rate leads to a decline in currency value, thereby affecting people’s purchasing power. When people anticipate a rising inflation rate, they tend to increase their consumption to avoid the future erosion of purchasing power.4 This leads to a decrease in investments and savings, thereby restraining economic growth. High inflation rates increase economic uncertainty and risk. During inflation, businesses face greater challenges in undertaking long-term planning and investment decisions owing to unstable price fluctuations and difficulties in cost forecasting. This reduces business investment willingness, resulting in a negative impact on economic growth. High inflation rates can distort resource allocation within an economy. As the inflation rate increases, some businesses and individuals may engage in speculative activities rather than substantive production and investment. This irrational resource allocation can weaken the economy’s productive capacity and growth potential.5

We conducted robustness tests to ensure the accuracy of our threshold model test results. Firstly, we used a denser increment of 0.25 percentage points. Secondly, to avoid endogeneity issues in the model, we referenced the studies by Mubarik and Riazuddin (2005) and Fakhri (2011), we conducted a two-stage least squares (2SLS) analysis using the lag of inflation and foreign direct investment as instrumental variables. To identify the most suitable instrumental variables, we performed the Underidentification Test, Weak Identification Test, and Overidentification Test. Ultimately, only the lag of inflation and FDI met the necessary conditions to serve as valid instrumental variables. This methodological adjustment aims to enhance the comprehensiveness and reliability of our analysis, ensuring that our research more accurately reflects the complex relationship between inflation rates and economic growth. Tables 6, 7, and 8 display the corresponding results. According to our analysis, the threshold level of inflation remains at 2%, consistent with the results presented in Table 5. Moreover, the regression results of the threshold model confirm that inflation rates below 2% have a positive impact on economic growth, while rates above 2% have a significant negative impact. Thus, the robustness analysis further substantiates the nonlinear relationship between inflation rates and economic growth. These findings deepen our understanding of the relationship between inflation rates and economic growth, indicating the existence of threshold effects at different levels. When inflation rates are low, consumers and businesses anticipate future price increases, leading them to prefer current consumption and investment, thereby stimulating economic growth. However, when inflation exceeds the threshold, its negative impact becomes significant, possibly due to the adverse effects of inflation on economic stability and resource allocation.

2)Mundell (1965).

3)Stockman (1981).

4)Stockman (1981).

5)Mundell (1965).

V. Concluding Remarks

This study examined whether there is a threshold effect of inflation on economic growth during 1987 to 2021 in China. The estimated threshold model indicates a nonlinear relationship between China’s economic growth and inflation, with a threshold level of 2% for inflation in China’s economic growth. When the inflation rate is below the threshold, inflation has a significant positive impact on economic growth. However, when the inflation rate exceeds the threshold, it has a significant negative effect on economic growth. This can be understood as moderate inflation stimulating economic growth by encouraging consumption, investment, increasing corporate profits, and government revenue. However, when inflation is too high, inflation expectations may be exacerbated, leading to increased production costs, decreased purchasing power, and reduced investments, ultimately hindering economic growth. Additionally, Granger causality tests and variance decomposition analysis based on the VAR model indicate that there is a unidirectional causal relationship between inflation and economic growth. Economic growth is influenced more by its own volatility; the contribution of inflation to economic growth exhibits a trend of initially increasing and subsequently decreasing. However, this study has its limitations. The annual time series data used in the research only covers the period from 1987 to 2021, and the current annual data may not capture economic fluctuations and policy changes on a shorter time scale. Future research could consider using quarterly or monthly data to provide a finer temporal resolution and a more sensitive analysis of economic dynamics. Based on these findings, several policy recommendations can be drawn. First, it is crucial for the government to strike a balance between inflation and economic growth when formulating monetary policy and identify an appropriate inflation level that supports both sustainable economic growth and price stability. Maintaining moderate inflation within the threshold level of 2% is ideal. This requires the implementation of effective monetary policies to manage inflation pressure and expectations. Second, policymakers should closely monitor inflation dynamics and take timely measures to prevent inflation from exceeding the threshold level, as excessive inflation can have adverse effects on economic performance. Finally, efforts should be made to enhance the resilience and adaptability of the economy to mitigate the negative impact of inflation shocks and promote long-term economic stability.